Deep learning models using intracranial and scalp EEG for predicting sedation level during emergence from anaesthesia
Authors' conclusion
The combination of iEEG and scalp EEG better predicted sedation level than either modality alone.
Study context
- Population
- Patients emerging from anaesthesia after placement of intracranial electrodes for medically refractory epilepsyplus 5 validation patients
- Participants
- seven patients
- Comparison
- Combined vs iEEG only vs Scalp EEG only
- Outcome
- AUC for predicting OAA/S score 0-2 vs 3-5
Methodological detailsdeep learning model, independent validation cohort
- Deep learning model
- Independent validation cohort
Performance
As reported in the abstract
Prediction of OAA/S 0-2 vs 3-5
| Test or model | AUC | P |
|---|---|---|
| Combined | 0.795 | P<0.001 |
| iEEG only | 0.750 | P=0.02 |
| Scalp EEG only | 0.764 | P<0.001 |
AUC
Combined0.795
iEEG only0.750
Scalp EEG only0.764
P
CombinedP<0.001
iEEG onlyP=0.02
Scalp EEG onlyP<0.001
Further findings
- Independent validation, scalp EEG only (5 patients)
- 0.844 · AUC
Original publication
Deep learning models using intracranial and scalp EEG for predicting sedation level during emergence from anaesthesia. BJA Open. Published 12 October 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →